You would probably need something like Abaqus for that, but I've never used STAAD.

What would be involved in modeling a underground pipe for dynamic seismic? Are we talking about a time-history input force? Varying time-history accelerations depending on which joints along the length are being considered?

If you're looking for non-linear soil modeling then the answer would be no. Those programs are primarily for elastic materials.

I've got a lot of experience with both RISA and STAAD. I don't think either of them are well adapted to handle this type of problem. The problem with dynamically modeling underground pipe is that you have complex soil behavior. Damping and stiffness of the soil are complex properties. They are a function of the vibrating frequency rather than constants. The soil itself is a nonlinear material. If you really want to get into a dynamic analysis of underground structures, I'd suggest getting your hands on a copy of "Geotechnical Earthquake Engineering" by Steven L. Kramer. There are some good examples in there (although they are quite complex) of how to do dynamic ground response analysis. My first instinct however, is to find a way to simplify the problem using some sort of equivalent static analysis. How is everyone else in the industry handling this type of problem?

Here are a couple of pointers on carrying out such an analysis using STAAD.Pro. The soil can be simulated using spring supports. The stiffness of the soil would have to be constant to carry out a dynamic analysis in STAAD.Pro because although STAAD.Pro supports defining variable spring stiffness, as of now it is limited to static analysis only. For a dynamic analysis, STAAD.Pro does offer the flexibility to actually change the damping corresponding to the modal frequencies. You can either explicitly specify the damping corresponding to each mode or have the software calculate the damping based on the frequency of each mode.

Sye123,

You're describing an eigenvalue or modal analysis. I'm not sure that would work. I don't think you can use nonlinear compression only supports with dynamic analysis. Also, while there are assumptions STAAD can make in eigenvalue analysis to allow for variable damping, I don't think STAAD allows you to do the same for the stiffness. Dynamic soil behavior is all about impedance (variable stiffness and damping). I could be wrong, but I think you're stuck with a nonlinear inelastic time-history analysis - which STAAD doesn't handle.

DCBII,

The option I described is applicable not only to eigensolution but to dynamic analysis like time history as well. While defining the time history, you need to choose the modal damping option (MDAMP). However as I mentioned in my post, nonlinear soil stiffness cannot be defined in a dynamic analysis and you cannot define compression only in a dynamic analysis either.

i was asked if i can do this.. i said no.. although i have done seismic response spectra analysis for structure using risa, i dont think i can use similar approach for underground.. as josh plum said, risa cant do it. i'm sure staad cant as well..

A couple of points:

At least for reinforced concrete, the structure above ground will be at least as non-linear as the soil/pipe system below ground. If an analysis based on linear properties is considered suitable for the above ground structure, why isn't it suitable for the underground structure?

If a non-linear dynamic analysis is required there are a lot of options in between Risa/Staad and Abaqus. I know Strand7 will do this, and I am sure there are many others.

Doug Jenkins
Interactive Design Services
http://newtonexcelbach.wordpress.com/